Optimum Delivery Plan for Trucks in Urban Area under Environmental Restrictions Using Probedata

本研究旨在揭示基于探测数据的送货活动的基本特征，探测数据由一套安装在卡车上的探测设备（包括GPS和数据记录器）收集。利用这些探测数据，我们制定了数学模型，以制定环境限制下的卡车交付计划。从降低货车运营成本和改善城市环境的角度出发，提出了不同货车的配送方案，并进行了评价。本研究的主要成果归纳如下：首先，我们建立了数学模型，确定最佳的出发时间，使指定的交货时间和选择最佳交货路线，考虑的不确定性的旅行时间和装卸时间的基础上探测数据。同时，我们还推导出了影响公路网上货车路径选择的一些因素，建立了以揭示性因素为解释变量的非集计路径选择模型。其次，我们针对不同的货车（宅配货车和集装箱拖车）制定了两种类型的配送计划。1)提出了基于货物OD需求的宅配卡车配送中心的配置方案，并从提高卡车配送服务效率和减少负面环境影响两个方面进行了评价。2)利用路径选择模型，对公路收费政策对改善公路沿线环境的影响进行了分析。

This study aims to reveal the basic characteristics of delivery activities based on probe date which were gathered by a set of probe equipment (comprising GPS and data logger) mounted to trucks. With these probe data, we formulated mathematical models to lay out delivery plans for trucks under environmental restrictions. Then some delivery plans for different trucks were proposed and evaluated from the viewpoints of both reduction in truck operation cost and improvement of urban environment. The main findings of this study are summarized as follows:First, we built mathematical models for deciding optimum departure time to make the designated delivery time and choosing optimum delivery route by considering the uncertainty of travel time and loading/unloading time obtained based on probe data. We also deducted some factors influencing route choice by trucks on road networks and developed disaggregate route choice models using revealed factors as explanatory variables.Second, we made two types of delivery plans for different trucks (home delivery service trucks and freight container trailer trucks). 1) We proposed allocation plans of distribution centers for home delivery trucks under the cargo OD demand, and evaluated them from improvement of efficiency of truck delivery service as well as reduction in negative environmental impacts. 2) We estimated the effects of toll policy on changing trucks' (trailer trucks) running routes from highways in urban area, environmentally vulnerable area, to expressway in coastal area to improve roadside environment, using our route choice models.